Over the past decade the genetics of development and cancer have converged in the identification of signaling pathways that control embryo patterning and, when aberrantly regulated, lead to cancer. Typically, tumors form in tissues in which the pathways normally operate. Tumors arise due to oncogenic mutations in components of these pathways that lead to ligand-independent constitutive activity. A classic example is the canonical Wnt signaling pathway, an essential developmental pathway that is disrupted in several kinds of tumor. Mouse Wnt1 is a prototypical oncogene first identified when it was shown that activation of Wnt1 by integration of Mouse Mammary Tumor Virus led to cancer of the mammary gland in mice. Mutations in several components of the Wnt/beta-catenin pathway have identified both oncogenes and tumor suppressors in this pathway that lead, in particular, to colorectal cancer. Recently, we have shown that embryos lacking Wnt5a display severe growth defects in the developing gastrointestinal tract. We are currently examining the mechanisms underlying this phenotype and are assessing potential genetic interactions between Wnt5a and current mouse models of colorectal cancer. A recent serendipitous result arising from our genetic studies of Wnt5a signaling has led to the discovery that aberrant Wnt signaling may lead to cancers of the immune system, specifically T-cell lymphomas. A large-scale survival study has revealed that relatively late-onset lymphomas and thymomas arise in Tcf mutants older than 6 months of age, however when one copy of Wnt5a is removed, severe, early-onset lymphomas and thymomas arise as early as 3 months of age. These results suggest that Wnt5a has tumor suppressor activity and we are currently performing experiments to address this hypothesis. In collaboration with Jeff Rubin (NCI-Bethesda) we have characterized the expression of the secreted Wnt antagonist sfrp1 during mouse embryogenesis. Sfrp1 is highly expressed in the developing embryo including the colon and small intestine. Given that sfrps are negative regulators of the Wnt pathway and that aberrant activation of the Wnt pathway leads to cancers of the gut, we are interested in determining whether sfrps have tumor suppressor activity. To test for tumor suppressor activity genetically, we have generated loss of function alleles of sfrp1. Despite the strong, early developmental expression, Srp1 null homozygotes are viable and fertile with subtle defects in bone density. Aging studies indicate that mutant animals have normal lifespan and do not develop tumors. Overlapping expression of sfrp1 with other sfrp genes suggest genetic redundancy so we are currently generating double and triple mutants between sfrp1, sfrp2 and sfrp3. Finally, genetic interactions between animals lacking sfrp and mouse models of colorectal cancer are also being examined.